Br. J. Pharmacol. (1992), 107, 1068-1074

'."

Macmillan Press

Ltd, 1992

Lack of effect of potassium channel openers on ATP-modulated potassium channels recorded from rat ventromedial hypothalamic neurones A.J. Sellers, *P.R. Boden & 'M.L.J. Ashford Department of Pharmacology, University of Cambridge, Tennis Court Road, Cambridge, CB2 1QJ and *Park-Davis Research Unit, Department of Pharmacology, Tennis Court Road, Cambridge, CB2 1QJ 1 Single neuronal cells were freshly isolated from the ventromedial hypothalamic nuclei (VMHN) of the rat brain. Currents through ATP-modulated and large conductance (160 and 250 pS) calciumactivated potassium channels were recorded by the cell-attached and excised inside-out patch techniques. 2 BRL38227 (lemakalim; 30-90 pM) applied to the superfusing medium produced no change in firing rate of isolated glucose-receptive VMHN neurones in cell-attached recordings. 3 BRL38227, at concentrations of between 30-100 LM applied to the intracellular (cytoplasmic) aspect of inside-out patches, had no effect on the activity of ATP-sensitive K+ channels in the absence of ATP or in the presence of a sub-maximal inhibitory concentration (3 mM) of ATP. Cromakalim, pinacidil, minoxidil sulphate and diazoxide also produced no effect under these conditions. 4 The potassium channel openers (KCO's) were tested on ATP-activated potassium channels recorded from a further subpopulation of VMHN neurones. Application of BRL38227 (up to and including 100 JAM) to this channel in inside-out patches either in the absence of ATP or when activated by 5 mM ATP had no effect on channel activity. Identical results were obtained with cromakalim and pinacidil. 5 BRL38227 had no effect on either of the large conductance (250 pS and 160 pS) calcium-activated potassium channels in VMHN neurones. 6 Intracellular recordings were made from glucose-receptive VMHN neurones in rat brain slices. Cromakalim (50 JAM) or diazoxide (60 pM) did not alter the firing rate or passive membrane properties of these neurones demonstrated to be sensitive to tolbutamide (0.1 mM). 7 These results show that the KCO's tested in this study have no effect either on VMHN neurones contained in brain slices or on the activity of any of the ATP-modulated potassium channels under isolated patch conditions associated with these neurones. Keywords: Potassium channel openers; BRL38227; ATP-sensitive K+ channels; Ca2"-activated K+ channels; hypothalamic neurones

Introduction The heterogeneous class of drugs known collectively as the potassium channel openers (KCO's), have been widely researched as powerful smooth muscle relaxing agents with both hypotensive and bronchodilator activity in vivo (for review see Robertson & Steinberg, 1990). Recently it has been suggested that these compounds may be of some use in diseases of the central nervous system (Miller, 1990; Aronson, 1992). Cromakalim (BRL 34915) is an isomeric benzopyran molecule with activity residing mainly in the 3S-4R configuration (BRL 38227, [lemakalim]). In the CNS, cromakalim has been shown to be effective in reducing seizure activity in two models of epilepsy (Gandolfo et al., 1989a,b), and has also been shown to affect neuronal excitability (Alzheimer et al., 1988; Politi & Rogawski, 1991). Furthermore, potassium channel openers have been shown to block neurosecretion (Schmid-Antomarchi et al., 1990). The target channel for these agents has been identified in some peripheral tissues. For instance, in cardiac muscle cromakalim has been shown to activate a glibenclamide-sensitive potassium current in isolated ventricular myocytes (Escande et al., 1988; Sanguinetti et al., 1988), and adenosine 5'triphosphate (ATP)-sensitive K+ channels isolated in membrane patches from the same cells have also been shown to be activated by cromakalim (Escande et al., 1988) and

'Author for correspondence.

pinacidil (Fan et al., 1990). In skeletal muscle cells, potassium channel openers including cromakalim, pinacidil and RP 49356 have also been demonstrated to activate a sulphonylurea-sensitive potassium conductance (Quasthoff et al., 1989). However, it is not as yet clear which type of potassium channel is affected by these drugs in the CNS (Alzheimer et al., 1988) or smooth muscle (Weston, 1989). Although it has been little used in CNS preparations the direct approach of single channel recording offers the best opportunity to determine the target site (channel) of these compounds and perhaps indicate the possible mechanisms of action. Single channel recording techniques have identified three distinct potassium channels in neurones in the ventromedial hypothalamic nuclei (VMHN) the activity of which is modulated by intracellular levels of ATP. The ATP-sensitive K+ channel is inhibited by increasing levels of intracellular ATP or by the sulphonylurea tolbutamide and is located in glucose receptive neurones where it contributes to resting membrane potential (Ashford et al., 1990a,b). This channel closes under conditions of elevated extracellular glucose (Ashford et al., 1990a) causing membrane depolarization in a manner analogous to that demonstrated in the pancreatic Pcell (Ashcroft et al., 1988). An ATP-activated K+-channel which in recordings from inside-out patches responds to elevated intracellular ATP levels with a large increase in channel activity, can be isolated from glucose sensing neurones which respond to elevated extracellular glucose with

KCO's AND ATP-SENSITIVE K+ CHANNELS

a hyperpolarization possibly mediated by the opening of these channels (Rowe et al., 1992). A third type of ATPsensitive K+ channel is sometimes observed in patches excised from VMHN neurones. This channel has been classed, according to its conductance (250pS), as a maxicalcium-activated potassium channel. However, this channel also shows sensitivity to changes in intracellular ATP (Treherne & Ashford, 1991). A separate large conductance (160pS) Ca2"-activated K+-channel has also shown to be present in glucose-receptive neurones of the VMHN, and its activity is not sensitive to intracellular ATP levels (Treherne & Ashford, 1991). As the ATP-K' channel is the target channel for cromakalim in cardiac muscle (Sanguinetti et al., 1985) and various potassium channel openers can, at high concentrations, inhibit insulin release from the pancreatic P-cell by increasing ATP-sensitive potassium channel activity (Garrino et al., 1989) the aim of the present experiments was to examine the effects of various potassium channel openers and this benzopyran molecule in particular on the membrane potential and spontaneous action potential firing rate of intact VMHN neurones and on the ATP-modulated potassium channels that can be recorded from isolated membrane

patches.

Methods

Electrical recording and analysis Coronal slices (350 gm thick) of hypothalamus were cut from brains of male Sprague-Dawley rats (80-200 g weight) with a Vibratome (Oxford Instruments). The slices were maintained at room temperature in artificial cerebrospinal fluid (ACSF) bubbled with 95% 02 and 5% CO2. For intracellular recording, the slices were transferred to a recording chamber where they were superfused with ACSF at 370C. Electrodes were filled with 1 M potassium acetate and had d.c. resistances of 100-150 MCI when measured in physiological saline. A period of 30 min was allowed for equilibration following impalement. Input resistances were derived from the slope of the current-voltage plot obtained by measuring the electronic potential during current injection. Pulses, of greater than 100 ms duration, were applied in order to ensure complete capacitance saturation of the membrane. All experiments were performed at 370C. These recording procedures have been described in full elsewhere (Boden & Hill, 1988). For single channel recording from VMHN neurones, cells were acutely dissociated from the VMHN isolated from rat hypothalamic slices. The nuclei were incubated with 0.5 mg ml' collagenase (Clostridiopaptidase A. Boehringer, Mannheim) and 1 mg ml' trypsin (Type XII, Sigma, Poole, Dorset) in ACSF at room temperature for 1-2 h and then triturated by the use of flame polished Pasteur pipettes. The dispersed cells were transferred onto Falcon 3001 dishes (35 mm) and left for 30 min to adhere. Single channel currents were recorded, at room temperature (21°C), from cellattached and inside-out membrane patches by standard patch-clamp recording procedures (Hamill et al., 1981). Current recordings were made with an Axopatch 2D patch clamp amplifier and stored on magnetic tape (Racal 4DS) for later reproduction of figures and analysis. The potential across the membrane is described with the usual sign convention for membrane potential (i.e. inside negative). The data were analysed for current amplitude and open-state probability P0p1. by computer (Apricot XEN-i 286/45) as described previously (Sturgess et al., 1988; Kozlowski et al., 1989). All data in the text are presented as mean values ± s.e.mean.

1069

Solutions The ACSF contained (in mM): NaCl 128.0, KCl 5.0, NaH2PO4 1.2, CaCI2 2.4, MgCl2 1.3, NaHCO3 26.0, D-glucose 10.0, pH 7.4. Before single channel recordings the cells were washed with normal physiological salt solution (PSS) consisting of (mM): NaCl 135.0, KCl 5.0, CaCl2 1.0, MgC12 1.0, HEPES 10.0 pH 7.4 with NaOH. For cell-attached and inside-out recordings the patch pipette contained (mM): KCl 140.0, CaCI2 1.0, MgCl2 1.0, HEPES 10.0, pH 7.2 with KOH and the bathing solution was either the normal external PSS (cell-attached recordings) or an intracellular solution (inside-out recordings) containing (mM): KCI 140.0, MgCl2 1.0, CaCl2 0.9, EGTA 1.0, HEPES 10.0, pH 7.2 with KOH (free Ca21 concentration of 0.8 JM). Free calcium concentrations were controlled and changed by use of EGTA, and determined by the 'METLIG' metal ion/ligand binding programme (P. England & R. Denton, University of Bristol).

Drugs All potassium channel openers were made up in a 10-2 M stock solution in 70% ethanol. Diazoxide, minoxidil sulphate and tolbutamide were obtained from Sigma Chemicals, Poole, Dorset. For brain slice experiments cromakalim was made up as a 10-2 M stock in dried dimethylsulphoxide (DMSO) and diluted to the required concentration in ACSF. We acknowledge kind gifts of pinacidil from Leo Pharmaceuticals, Ballerup, Denmark and cromakalim and BRL38227 from SmithKline Beecham Pharmaceuticals, Welwyn, Herts.

Results Inside-out patches were obtained from isolated VMHN neurones bathed in symmetrical potassium solutions. In approximately one third of patches obtained from the VMHN cells (n = 28) the ATP-K+ channel was observed and identified by its conductance (146 pS) and sensitivity to ATP. The ATP-K+ channel can be isolated from glucose-receptive neurones that respond to elevated extracellular glucose with a depolarization mediated by closure of these ATP-K+ channels (Ashford et al., 1990a). ATP caused a concentrationdependent reduction in channel activity with an IC50 of approximately 3 mm, with channel activity exhibiting no sensitivity to calcium (Ashford et al., 1990a). To reduce the activity of the large conductance (160 pS) calcium-activated potassium channel observed in most patches, the free calcium concentration was reduced to very low (nM) levels. Figure 1 shows a typical trace showing activity of an ATP sensitive K+ channel recorded from an inside-out patch isolated from a VMHN neurone. In the absence of ATP the channel had a Popen of 0.863. The addition of 3 mM ATP caused a reduction in channel activity to an open probability of 0. 110. BRL38227 was then applied concomitant with 3 mM ATP at a concentration of 30 gM and caused no observable change in channel activity (Popen; 0.110). The effects of ATP were readily reversed on washing (Ppen; 0.870). BRL38227 was also applied under identical conditions at a concentration of 100 JM in three further experiments and no activation of channel activity was observed. Likewise, no other potassium channel opener tested on ATP-K+ channel activity recorded from inside-out VMHN membrane patches, inhibited submaximally by 3 mM ATP, produced any observable activation. The openers tested were cromakalim (100 JM, n = 2), pinacidil (30 JM, n = 2, and 100 JM, n = 2), minoxidil sulphate (100 JM, n = 3) and diazoxide (100 JM, n = 2). For example, the typical open state probability of the ATP-K+ channel in such an experiment was; control 0.774, 3 mM ATP 0.250, 3 mM ATP + 100 JM pinacidil 0.246, wash 0.800. In order to determine whether the KCO's had an underlying effect on ATP-K+ channel activity, they were applied to the isolated inside-out patch in the absence of ATP. For exam-

A.J. SELLERS et al.

1070

5 mM ATP

a

4d

b

4j1-11133

CL b

d

'E". 5pAL 2s Figure 1 Single channel currents recorded from a single inside-out patch, excised from a VMHN neurone, held at a membrane potential of + 30 mV. The recording pipette contained (in mM). KCl 140, MgCl2 1 and CaCl2 1, and the bath (in mM) KCI 140, MgCI2 1, EGTA I (with a free calcium concentration of

Lack of effect of potassium channel openers on ATP-modulated potassium channels recorded from rat ventromedial hypothalamic neurones.

1. Single neuronal cells were freshly isolated from the ventromedial hypothalamic nuclei (VMHN) of the rat brain. Currents through ATP-modulated and l...
1MB Sizes 0 Downloads 0 Views